isprsarchives-xxxviii-4-c26-1-2012

8/13/2019 isprsarchives-XXXVIII-4-C26-1-2012

1/5

INDOOR LOCALIZATION USING WI-FI BASED FINGERPRINTING ANDTRILATERATION TECHIQUES FOR LBS APPLICATIONS

Solomon Chan 1*, Gunho Sohn 1 1GeoICT Lab; Earth and Space Science and Engineering Department, York University, Toronto Canada

(sol122, gs ohn)@yorku.ca

KEY WORDS: Indoor Localization, Access Points, Trilateration, Fingerprinting, Database, Wi-Fi, RSS, GIS

ABSTRACT:

The past few years have seen wide spread adoption of outdoor positioning services, mainly GPS, being incorporated into everydaydevices such as smartphones and tablets. While outdoor positioning has been well received by the public, its indoor counterpart has

been mostly limited to p rivate use due to its higher costs and complexity for setting up the proper environment. The objective of thisresearch is to p rovide an affordable mean for indoor localization using wireless local area network (WLAN) Wi-Fi t echnology. Wecombined t wo different Wi-Fi approaches to locate a user. The first method involves the use of matching the pre-recorded receivedsignal strength (RSS) from nearby access points (AP), to the data transmitted from the user on t he fly. This is commonly known as

fingerprint matching . The second approach is a distance-based trilateration approach using three known AP coordinates detectedon the users device to derive the position. The combination of the two steps enhances the accuracy of the user posit ion in an indoorenvironment allowing location-based services (LBS) such as mobile augmented reality (MAR) to be deployed more effectively in theindoor environment. The mapping of the RSS map can also prove useful to IT planning personnel for covering locations with noWi-Fi coverage (ie. dead spots). The experiments p resented in this research helps provide a foundation for the integration of indoorwith outdoor positioning to create a seamless transition experience for users.

1. INTRODUCTION

Location-based services (LBS) require on-t he-fly localizat ion ofthe user to provide its service and other information. Indoorlocalization applications, such as indoor navigation, require

precise user location. Since p ositioning services such as GPSremain inefficient for indoor use, other sensor techniques should

be considered. With t he maturation in W i-Fi t echnologies andthe increasing capabilities of hardware in our smart devices,locating a person indoor became a good starting point for

providing indoor LBS.

This abstract uses two different Wi-Fi based approaches tolocate a person in an indoor space. First s tep requires the use offingerprint matching to compare signal strength data receivedfrom nearby access points (AP) by the user, to the referencedata stored in the Wi-Fi signal strength database. The secondmethod is by means of trilaterating the distances between threeknown AP coordinates thereby calculating the distance the useris relative to an AP (see Figure 1).

Figure 1 Steps performed to compute a users position

In this abstract, a general overview of an indoor localizationsystem, currently under development, will be presented which isintended to be incorporated into a 3D v irtual campus model thathas been demonstrated previously in the 6 th 3D Geoinfo [6] asshown in Figure 2. Such a system can dynamically representreal world information onto a virtual world thereby increasingthe usefulness of 3D semantic data models.

Figure 2 Indoor 3D models can be a useful aid forvisualizing real-time dynamic information such as trackinga moving person [7]

2. MOTIVATION & BACKGROUND

Building on last years success on the integration of dynamic3D virtual world and tracking moving objects using surveillancevideo camera, the focus for our group now shifts towardlocalizat ion using Wi-Fi techniques. We would like to explorethe feasibility and compare the accuracy of a visual video based

approach for indoor tracking versus a more generalized Wi-Fi based approach.

1.Measurements fromobserversent to

database

2. Performfingerprintmatching

3. Applytrilateratio

nalgor ithm

IndoorLocalization

International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXVIII-4/C26

7th International Conference on 3D Geoinformation, May 16-17, 2012, Quebec, Canada 1 of 5


2/5

There are many technologies that can be used for indoorlocalization and posit ioning. Radio Frequency ID (RFID),Ultrasound, Infrared Beacons, Bluetooth, Global System forMobile communication (GSM), Wireless Local Area Network(Wi-Fi), just to name a few. Depending on location and need,one option will be better than another. For instance, using Wi-Fi based methods in a hospital environment could prove

dangerous as the Wi-Fi signals could interfere with hospitalequipment radio frequency signals. On the other hand, densely

placed Wi-Fi AP will provide a solid coverage in most urbansett ings. Additionally with the number of Wi-Fi capabledevices in the market and an abundance of pre-existing Wi-Fiinfrastructure at present time, Wi-Fi has become a logicalchoice to use for tracking and providing LBS to general public.

Recent research such as [7] demonstrates the feasibility ofindoor localization and has even been applied into an indoornavigational context operating on an Android device. However,the p recision of the user location is not very well explored as thefocus of the research was on traversing from point A to B. Assuch, it is more of a symbolic representat ion of locations. It is,

however, a useful first approximat ion of the users location. Bolliger [1], on the other hand, explored the issue oflocalization accuracy using multiple deterministic and

probabilistic methods based on WLAN fingerprinting and testedthe feasibility of crowdsourcing to improve the radio map

precision in the database. In fact, Bolliger has recentlydeveloped an indoor positioning system named Redpin thatallows user to voluntarily up load their location to their server tohelp contribute and enhance the accuracy of their positioningsystem. By increasing the location point density, the fingerprintmatching can also become more precise. And while multiplemethods have been considered, they are all fingerprint matchingrelated.

2.1 Positioning VS Localization

A clear distinction should be made between the terms positioning and localization. Positioning is the determination ofglobal world coordinates (ie. 43.77568, 77.12243) whilelocalization is the determination of relative coordinates (ie.Petrie Science and Engineering Building, Room 101) [4].

Positioning is particularly important for trilateration method ascoordinates must be known for access points prior to anydistance determination. On the other hand, fingerprintingapproach is much more reliant on symbolic representation andthus does not require absolute positioning to be known.

2.2 FingerprintingWi-Fi fingerprinting requires a robust RSS database which will

be used for generating signal strength maps as well as used formatching. Each reference point includes signal strengthmeasured from all accessible AP. Live RSS data can then becompared to the find the closest match from the database whichstores the location of each reference point [5].

Fingerprint matching algorithm generally consists of twocomponents: the radio map and the estimation method. Theradio map must be established as part of the training phase to

building up the database. The most commonly used estimationmethod, and also used in this experiment, is the Nearest

Neighbour method. Other methods include the Support VectorMachine, as well as Hidden M arkov Model.

2.3 Trilateration

Trilateration method uses distance from nearby AP with knownMedia Access Control (MAC) addresses, calculated from signalstrength values, to approximate the distance to the user. It isimportant to note that different networks of AP, particular oneswith different hardware configurations, may vary in calculating

the distance. Unlike fingerprinting, this method does notrequire priori data collection [8].

2.4 Signal Multipath Propagation

In any study involving indoor Wi-Fi signals, one must considerthe effect of multipath propagation (ie. reflection and absorptionof signals). Due to the nature of multipath, it is important toconsider a range of values when trying to match the signalstrength with the pre-recorded database values.

3. EXPERIMENTAL SETUP

To perform localization, a resourceful database with accuratecoordinates is needed. Additionally, due to the mult ipathinterference of Wi-Fi signals, it is unrealistic to expect a 100%

perfect match in signal between the database and the live feed insignal from the user. Therefore, each reference point uses anaverage RSS value taken from ~15 seconds worth of signaldata. To ensure the effect of multipath fluctuation is minimized,a range of RSS values will be considered as possible candidates.The range will be selected based on the standard deviation ofthe raw data (ie. standard deviation of ~15 seconds of signalstrength values). Trilateration calculations will be computed

based on RSS from nearby AP by considering p ath loss (t o bediscussed in section 3.4 below).

3.1

Data Collection

The reference data to be used by the database for comparison inthe fingerprint matching process consists of 84 locationsdistributed evenly on floor 1 to 3 of Petrie Science andEngineering (PSE) Building at York University . The RSS aremeasured in units of dBm, using the open source softwareinSSIDer2.0, along with its corresponding MAC address, SSID,channel, etc. inSSIDer2.0 is a freeware used to detect wirelesssignals in multiple channels and it has been used in previousresearches [5] p roviding accurate data.

All AP in the PSE building are the same Cisco AP1120B modelthat supports 802.11b band which operates at 2.4GHz rangeonly. The AP are transmitting at full strength at 100mW(corresponds to 20dBm).

3.2 S ystem Requirements

The preliminary system is aimed toward Android users due toits large user base and open source system development . Bothsmartphone and tablets with an Android OS can install anapplication that allows RSS data from the device be sent to thedatabase for analysis, thereby computing the users indoorlocation.

3.3 Radio Map

The CAD floor plans of the PSE building was geo-referenced in

ArcGIS as p art of the AP mapping process which is required forgenerating the RSS map. See Figure 3 for a RSS map of the 3 rd floor of the PSE buildin g. Note that signal strength




3/5

measurements were taken in corridors; therefore no empiricalRSS data were plotted for the offices and classrooms at theouter ring of the building at this point.

Figure 3 Received Signal Strength (RSS) radio map forcorridors on 3 rd floor of Petrie Science and EngineeringBuilding at York U ni versi ty

3.4 Distance Computation

Computing the distance between the user and a nearby APrequires knowledge of the signal path loss which can generally

be represented using the Log-distance Path Loss equation [2]:

(1)

where PL = path loss between sender and receiver = path loss at a distance of 1 meter away

d = distance between sender and receivern = path loss exponent in environments = standard deviation of signal strength variability

Distance d from a specific AP can be computed using theabove equation. Note that the distance comput ed represents aradius d meters around the AP to the user. By consideringthree nearest distances, trilateration can t hen be app lied to solvefor the intersection point where the user would be located. Anoverview of the system is demonstrated in the flow chart inFigure 4.

Figure 4 Sequential flowchart presenting the logic of thesystem

4 PRELIMINARY RESULTS

4.1 Fingerprinting Approach

An unknown user position was tested in this method asdemonstrated below. The user received signals from a list ofAP, each with their unique MAC addresses (see Table 1).

MAC Avg RSS

00:40:96:A1:0D:3A -33.15

00:40:96:A1:0C:B1 -56.62

00:40:96:A1:0C:AD -59.69

00:02:8A:9E:4F:31 -62.77

00:40:96:A1:0D:08 -64.77

00:40:96:A1:0D:45 -66

00:40:96:A1:0D:7C -67

00:09:B7:7B:9F:60 -86

00:40:96:A1:0C:A3 -86

Table 1 List of signal strengths (dBm) measured at oneunknown location from user

This unknown position with the given list of unique MACaddresses is then compared to the master list in the database thatstores all detectable MAC addresses within the building. In this

particularly case, it can be concluded t hat the user is at posit ionP0301, as evident from Table 2, which can then be translated

from a symbolic point t o a real world reference point.

Position ID

# ofMatches

Position ID

# ofMatches

Position ID

# ofMatches

P0101 2 P0201 0 P0301 8

P0102 1 P0202 0 P0302 1

P0103 0 P0203 0 P0303 1

P0104 0 P0204 0 P0304 0

P0105 2 P0205 0 P0305 0

P0106 0 P0206 1 P0306 0

P0107 0 P0207 2 P0307 0

P0108 1 P0208 0 P0308 0P0109 1 P0209 1 P0309 3

RSS values in dBm

- 9 9 . 7 7

- - 7 5

- 7 4 .

9 9 -

- 7 0

- 6 9 .

9 9 -

- 6 5

- 6 4 .

9 9 -

- 6 0

- 5 9 .

9 9 -

- 5 5

- 5 4 .

9 9 -

- 5 0

- 4 9 .

9 9 -

- 4 5

- 4 4 .

9 9 -

- 4 0

- 3 9 .

9 9 -

- 3 5

- 3 4 .

9 9 -

- 2 9 .

4 3




4/5


5/5

University of Saskatchewan, Canada.http:// rose.geog.mcgill.ca/ski/system/files/fm/2011/Wei.pdf (2June, 2011)


http://rose.geog.mcgill.ca/ski/system/files/fm/2011/Wei.pdfhttp://rose.geog.mcgill.ca/ski/system/files/fm/2011/Wei.pdfhttp://rose.geog.mcgill.ca/ski/system/files/fm/2011/Wei.pdf

isprsarchives-xxxviii-4-c26-1-2012

Documents